Building SYMLOG profiles with an online collaborative game

Collaboration is essential for individuals working in groups to achieve a common goal. The understanding of the collaborative profile of each member of a team is extremely useful to understand and predict his/her performance in future teamwork. Users can demonstrate their collaborative skills on many digital platforms. Among them, video games enable to capture the players' behavior by the direct observation of their actions, while engaging them in a pleasant activity. In this work, we propose an approach for building collaborative profiles of a group of people working together towards a common goal, using an online game as a shared environment and a well-known theory about groups' dynamics: SYMLOG. This profile can be useful to know which features each member should train to improve his/her collaborative skills and to predict the performance of the group. We validate our approach with 98 players to evaluate the similarity between the profiles generated by our approach and the profiles derived from the SYMLOG questionnaire, which is the usual tool used with this theory